There is a network system that employs a master-slave method for the field of instrumentation. FIG. 1 is a diagram illustrating an example of the configuration of the network system that employs the master-slave method (system-switching system 500).
The network system illustrated in FIG. 1 is composed of two masters 100 (control-system master 100a, standby-system master 100b) and a plurality of slaves 200 (slave 1 (2001), . . . , slave n (200n)) in order to maintain availability. Maintaining availability means to shorten the time during which the system is halted as much as possible.
By providing the duplicated masters 100 as described above, the standby-system master 100b can take over the role even if a malfunction occurs in the control-system master 100a, and availability can be maintained.
In order to maintain availability, it is necessary not only to provide the duplicated masters 100 but also to reduce down-time which occurs when the role of the control-system master 100a and the role of the standby-system master 100b are switched.
The down-time is the time during which a system or service is halted, and refers herein to the time from when a malfunction occurs in the control-system master 100a to when communication is resumed between the standby-system master 100b and the slave 200.
As an example of a technique for reducing the down-time, Patent Literature 1 is presented. Patent Literature 1 reduces the start-up time of an execution on-line system of a standby computing machine when a failure occurs in an execution computing machine. A system of Patent Literature 1 is composed of one execution computing machine and one standby computing machine. An execution on-line system is incorporated in the execution computing machine. A dummy on-line system is incorporated in the standby computing machine. A switching monitor program is incorporated in each of the execution computing machine and the standby computing machine.
The operation of the system of Patent Literature 1 is as described below. First, the execution computing machine operates at normal times. At this time, the dummy on-line system that is identical with the on-line system of the execution computing machine is launched on a memory in the standby computing machine. Thereafter, the dummy on-line system is executed on the standby computing machine until switching of the systems occurs.
When a failure occurs in the execution computing machine, switching of the systems occurs and the processing of the execution computing machine is taken over by the standby computing machine. In the standby computing machine, the identical dummy on-line system has already been started, so that the processing can be continued without initializing the memory, and the down-time can be reduced.